Skip to content
AI Work Index

V4.3 Shadow Model Note

Status: V4.3-shadow task-weighted shadow model

Shadow published

This page explains release impact and readiness. It does not replace the formula spec on Methodology, the threshold detail on Appendix, the schema/download contract on Data, or the citation layer on Research.

Shadow baseline

V4.2

historical comparison basis for this shadow artifact

Required inputs ready

4/4

present locally for shadow scoring

Published task-native occupations

492

occupations currently scored with the task-native shadow model

Validation comparison

2/3

current match-or-improve gates passing

Published shadow artifacts

The shadow layer is now auditable as data, not just as a readiness note.

Current shadow validation deltas: cluster -0.6667, BLS -0.0283, family -0.2143.

What Changes Already Affect Users

Current live V7 keeps this separate

  • Bootstrap uncertainty intervals are published on occupations in the live dataset.
  • Structural risk and near-term risk are separated in the forecast layer.
  • Task-primitives fields now publish weighted evidence where normalized O*NET task matches exist; sparse occupations remain explicit null.
  • The release and governance surfaces now expose shadow-model readiness instead of hiding it.
  • 492 occupations have published task-native shadow scores for archived comparison.

What still does not affect the headline score

  • No implausible anchor label flips without written rationale: still pending review

Remaining Input Gaps

  • All required local shadow-model inputs are now present.

Input Readiness

anthropic task penetration

data/raw/external/anthropic_task_penetration.csv

present

onet task statements

data/raw/external/onet/Task_Statements.txt

present

onet task ratings

data/raw/external/onet/Task_Ratings.txt

present

empirical mobility

data/raw/external/sg_empirical_mobility.json

present

Coverage Snapshot

Occupations

562

current published universe

Direct mapped

508

eligible for the direct coverage gate

Median direct matched task share

100%

current direct-coverage gate basis

Task-weighted share

88%

archived outside the live V7 headline

Promotion Gates

GateThresholdActualState

Median matched task weight share across direct-mapped occupations

This gate prevents a sparse task layer from directly changing the headline score before task matching is broadly comparable.

>= 0.61pass

Experimental task-adjusted score matches or improves current validation diagnostics

Requires at least 2 of 3 external checks to match or improve baseline. Current results: cluster directional accuracy 0.3333 vs 1; BLS rho -0.167 vs -0.1387; family rho -0.4633 vs -0.249.

at_least_2_of_32/3pass

No implausible anchor label flips without written rationale

8/8 anchors screened; 1 candidates still need editorial sign-off.

zero_unexplained_flips1pending

If V4.3 is eventually promoted

The intended direction is a task-weighted shadow model with effective coverage, automation-pressure, augmentation-upside, and concentration-aware net risk. Those candidate formulas are published as governance scaffolding, not live scoring rules.

effective_coverage = Σ_t w_it · exposure_t · success_t

net_risk = automation_pressure_i · (1 - λ · concentration_i) · market_modifier_i

What Must Happen Next

  • Review the published shadow artifact against the current validation misses.
  • Review anchor occupations and document any surprising label flips before promotion.
  • Keep the current score published until the shadow model matches or improves current validation.
  • Treat the empirical mobility prior as supporting evidence until a higher-granularity Singapore transition dataset exists.